This study asked three questions of inpatient neurology: (1) How often are routine overnight vital signs abnormal? (2) How often are those abnormal vital signs clinically significant? (3) Are there low-risk patients who may not need routine overnight vital signs?

We analyzed retrospective data from electronic health records. We identified routine vital sign measurements and quantified how often they were abnormal. We measured how often urgent clinical actions (stat orders, rapid responses, or transfers to higher level of care) occurred within 1 hour after vital sign checks.

We analyzed 102,184 routine vital sign events from 5,569 admissions to non-ICU neurology services over 7 years. In total, 12% of vital sign events were abnormal. Compared to after normal vital signs were obtained, the likelihood that an urgent clinical action was taken increased after abnormal vital signs were procured during the day in all patients (4.1% vs 2.4%, p-value<0.001); at night in all patients (2.2% vs 0.9%, p-value<0.001); and at night in patients with normal daytime vitals (2.0% vs 0.7%, p-value<0.001). The number needed to treat (NNT), signifying the number of vital sign events needed to generate one additional urgent clinical action, was 473 during the day, 635 at night, and 1,463 at night in low-risk patients with normal daytime vital signs. Stroke and non-stroke patients were similar, but the low-risk subgroups differed in their NNT (stroke NNT 1,168, non-stroke NNT 1,882).

Routine vital signs that are abnormal and warrant urgent clinical intervention are uncommon in non-ICU neurology inpatients. This is especially true at night and in low-risk patients, questioning the utility of checking overnight vitals, particularly when weighed against the impact of disrupting patients’ sleep.